Loss calculation method.

Method to create embedding using W, H, T and the node ids.

Initialization of weights and loss container.

Method to create adjacency matrix powers, read features, and learn embedding. :param args: Arguments object.

Function to print the losses in a nice tabular format.

Method to calculate a sparse degree normalized adjacency matrix. :param graph: Sparse graph adjacency matrix. :return A: Normalized adjac

Gradient descent updates for a given number of iterations.

A method to parse up command line parameters. By default it gives an embedding of the Wiki Chameleons. The default hyperparameters give a goo

Method to get dense node feaures. :param feature_path: Path to the node features. :return features: Node features.

Method to read graph and create a target matrix by summing adjacency matrix powers. :param edge_path: Path to the ege list. :param order:

Method to get sparse node feaures. :param feature_path: Path to the node features. :return features: Node features.

Saving the embedding on disk.

Function to print the logs in a nice tabular format. :param args: Parameters used for the model.

Feature embedding matrix update method.

Node emebdding matrix update method.

Setting up the target matrices and arguments. Weights are initialized. :param A: Proximity matrix. :param T: Text data

Calculating the losses in a given iteration. :param iteration: Iteration round number.

Calculating the losses in a given iteration. :param iteration: Iteration round number.

Saving the embedding on disk.

Saving the embedding on disk.

A single update of the feature basis matrix.

A single update of the feature basis matrix.

A single update of the node embedding matrix.

A single update of the node embedding matrix.